DE69331194T2 - ADENO ASSOCIATED VIRUS WITH REVERSE TERMINAL REPEAT SEQUENCES AS A PROMOTOR FOR THE TRANSFER OF A FUNCTIONAL CFTR GENE IN VIVO - Google Patents

Abstract

Described herein are constructions of recombinant DNA comprising modified adeno-associated virus (AAV) DNA sequences capable of functioning as a eukaryotic expression vector for expressing foreign DNA sequences using a novel transcription promoter comprising the termini of AAV DNA. It is shown that expression of a test reporter gene can be obtained from this vector in mammalian cells. It is further shown that this combination of vector and promoter can be used to introduce and express a human gene and correct a genetic defect in human cells resulting from malfunction of the mutant endogenous gene. Further, the vector can be used to correct the genetic defect by expressing a modified version of the human gene consisting of a fusion of part of the said gene and a synthetic sequence contained in the vector. <IMAGE>

Description

Background of the invention

Adeno-associated virus (AAV) is typically defective in replication and depends on coexisting adenovirus or herpesvirus infection for effective replication and a productive life cycle. In the absence of a helper virus, AAV can undergo stable integration of its genome into the host cell; however, the integrated AAV genome has no pathogenic effect. These properties provide the basis for the development of AAV vectors for gene expression in mammalian cells. AAV vectors are used to express both selective markers (Hermonat and Muzyczka, 1984, Proc. Natl. Acad. Sci. (USA) 81: 6466-6470; Tratschin et al., 1985, Mol. Cell. Biol. 5: 3251-3260), such as neomycin phosphotransferase (neo), and non-selected genes including chloramphenicol acetyltransferase (cat) (Tratschin et al., 1984, Mol. Cell. Biol. 4: 2072-2081) and thyroid-stimulating hormone in eukaryotic cells (Mendelson et al., 1988, Virology 166: 154-165; Wondisford et al., 1988, Molec. Endocrinol. 2: 32-39).

For use as a viral transducing vector, AAV may have several advantages including a high frequency of stable DNA integration and the lack of pathogenicity of wild-type AAV. One limitation regarding AAV is that of size, since the packaging limit for foreign DNA into AAV particles is approximately 4.5 kilobases. This limitation is an important consideration for the development of AAV vectors for the expression of genes or cDNA constructs in which the coding gene sequence approaches the AAV packaging limit, i.e., approximately 4.5 kilobases.

One such gene is the cystic fibrosis gene (CFTR). The airway epithelium is a critical site of cellular dysfunction in cystic fibrosis (CF), the most common fatal genetic disease in North America, and the disease is characterized by a defect in the regulation of Cl- conductance (Hwang et al., 1989, Science 244: 1351-1353; Li et al., 1988, Nature (London) 331: 358-360; Li et al., 1989, Science 244: 1353-1356; Schoumacher et al., 1987, Nature (London) 330: 752-754). The cDNA for the CFTR gene (Riordan et al., 1989, Science 245: 1066-1073; Rommens et al., 1989, Science 245: 1059-1065) was expressed in eukaryotic cells. Expression of CFTR protein in non-epithelial cell lines resulted in the formation of Cl- conductance (Andersen et al., 1991, Science 251: 679-682; Kartner et al., 1991, Cell 64: 681-691). The OF defect was complemented by expression of CFTR in a CF pancreatic adenocarcinoma cell line by stable transduction with a retrovirus vector (Drumm et al., 1990, Cell 62: 1227-1233) and in a CF airway cell line by infection with a vaccinia virus (Rich et al., Nature (London) 347: 358-363) or an adenovirus vector (Rosenfeld et al., 1992, Cell 68: 143-155).

Gene therapy has been proposed as a way to reverse the cellular defect and prevent disease progression in affected patients. Previous gene therapy approaches have involved in vitro transduction of cells (such as lymphocytes) that can be easily reintroduced into patients. This may be difficult in an intact airway epithelium. An alternative approach is to use a viral vector to deliver the CFTR gene directly to the airway surface. One candidate is adeno-associated virus (AAV), a human parvovirus. However, the coding sequence (Riordan et al., 1989, Science 245: 1066-1073) of CFTR is 4.4 kilobases, approaching the packing limit for AAV particles. AAV therefore has the potential disadvantage with regard to use as a vector for CFTR that it can hardly accommodate the coding sequence of CFTR (Collins, 1992, Science 256: 774-779).

Transducing AAV vectors are described in the patent to Carter et al., (US Patent 4,797,368, issued January 10, 1989). This patent describes AAV vectors using AAV transcription promoters p40, p19 and p5.

AAV vectors must have a copy of the AAV inverted terminal repeat sequences (ITRs) at each end of the genome in order to be replicated, packaged into AAV particles, and efficiently integrated into cell chromosomes. The ITR consists of nucleotides 1 to 145 at the left end of the AAV DNA genome and the corresponding nucleotides 4681 to 4536 (i.e., the same sequence) at the right end of the AAV DNA genome. Thus, AAV vectors must have a total of at least 300 nucleotides of terminal sequence.

To package large coding regions, such as the CFTR gene, into AAV vector particles, it is important to use the smallest possible regulatory sequences, such as transcription promoters and poly-A addition signal. Also in this latter study and another study (Beaton et al., 1981, J. Virol. 63: 4450-4454) it was shown that the AAV ITR sequence can serve as an enhancer for the transcriptional promoter of the SV40 virus early gene. However, the AAV ITR region was not shown to have any intrinsic transcriptional promoter activity. Indeed, the literature teaches that the AAV ITR regions have no transcriptional function (Walsh et al., 1992, PNAS (in press)). Therefore, previous AAV vectors have used a small transcription promoter, namely the AAV p5 promoter, which consists of nucleotides 145 to 268 of the AAV genome, located immediately adjacent to an ITR.

Summary of the invention

According to the invention, a novel functional cystic fibrosis transmembrane conductance regulator (CFTR) protein has a deletion of any or all of the amino-terminal 118 amino acids. Furthermore, a novel polynucleotide comprises the inverted terminal repeat (ITR) sequences of adeno-associated virus and a heterologous nucleic acid encoding the CFTR protein, wherein the ITR sequences serve as a promoter for transcription of the nucleic acid. A vector comprising the polynucleotide can be used in the treatment of cystic fibrosis, particularly in humans.

AAV vectors containing the full-length CFTR cDNA are larger than wild-type AAV and difficult to package into transducing AAV particles. However, a CFTR cDNA expressed from an AAV ITR promoter according to the invention can complement the CF defect and is appropriately regulated as shown by functional assays. It has been shown that this truncated CFTR cDNA can be packaged into an AAV vector and made to infect IB3 cells so that the CF defect could be complemented in culture. It is thus possible to achieve efficient complementation of the CF defect with AAV transfecting vectors.

Description of the invention

Preferably, the vector used in the invention is an adeno-associated virus vector. By "adeno-associated virus vector" is meant any vector that has the ITR sequences required for packaging of the viral genome, integration into a host chromosome, and promotion of transcription of additional sequences. Thus, any changes in the ITR that maintain these essential functions are considered to be within this scope.

The polynucleotide may also comprise a poly-A site capable of being read translationally in the reverse direction. The poly-A site comprises SEQ ID NO: 6.

The viral vector can be contained within a suitable host. Any cell can be a suitable host as long as the vector is capable of infecting the cell type. An example of a suitable host cell is an epithelial cell that contains a non-functional CFTR sequence.

The vector may contain additional sequences, such as from adenovirus, that help to achieve a desired function of the vector. For example, the addition of adenovirus DNA sequences encapsulating the AAV vector could provide one approach to packaging AAV vectors into adenovirus particles.

The vector may also be contained in any pharmaceutically acceptable carrier for administration. Examples of suitable carriers include saline and phosphate buffered saline.

As used here, AAV means all serotypes of AAV. It is routine practice in the field to use the ITR sequences from other serotypes of AAV, since the ITRs of all AAV serotypes are expected to have similar structures and functions in terms of replication, integration, excision, and transcriptional mechanisms.

Experimental procedures and results

Cells: The CFBE-IB3-1 cell line (IB3 cells) is a human bronchial epithelial cell line derived from a CF patient and immortalized with an adeno/SV40 hybrid virus (Luo et al., 1989, Pflugers Arch. 415: 198-203; Zeitlin et al., 1991, Am. J. Respir. Cell Mol. Biol. 4: 313-319). These cells retain characteristic properties of epithelial cells and are defective in Protein kinase A activation of chloride conductance. IB3 cells were grown at 37°C in 5% CO2 in LHC-8 medium (Biofluids, Inc., Md) plus 10% fetal calf serum with added endothelial cell growth supplement (15 μg/ml) in culture flasks or dishes coated with collagen (150 μg/ml), fibronectin (10 μg/ml), and bovine serum albumin (10 μg/ml). The 293-31 cell line (293 cells), originally derived from human embryonic kidney cells transformed with the adenovirus type 5 E1A and EIB genes, was grown at 37°C in 5% CO2. in Eagle minimal essential medium with 10% fetal calf serum and used for packaging of AAV vectors into virus particles (Tratschin et al., 1984, Mol. Cell. Biol. 4: 2072-2081).

Plasmids: Plasmids were constructed and grown using standard procedures (Sambrook et al., 1989, Molecular Cloning, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York). The starting plasmid pAV2 contains the entire 4681 nucleotide sequence of AAV2 inserted into a pBR322-derived plasmid via a polylinker and BglII linker (Laughlin et al., 1983, Gene 23: 681-691). This resulted in a plasmid pYT45 containing a prokaryotic cat gene immediately downstream of AAV nucleotides 1-263 (placing the cat gene under the control of the AAV p5 promoter), followed by AAV nucleotides 1882-1910 and 4162-4681 (containing the poly-A signal and the right ITR) downstream of the cat gene.

pAAVp5neo is analogous to pYT45 except that it has a neo coding sequence in place of the cat gene and the downstream AAV nucleotides 1882-1910 and 41624492 (the KpnI/SnaB fragment) have been replaced by 60 bp SPA.

pSA313 is analogous to pAAVp5neo except that the neo sequence was replaced by the CFTR coding sequence contained in a 4502 bp AvaI-SstI fragment excised from a plasmid pBA-CFTRBQ (Drumm et al., 1990, Cell 62:1227-1233). This CFTR cDNA sequence contains three silent mutation sites in exon 6a that eliminate the prokaryotic promoter sequence. In pSA313, the CFTR gene is under the control of the AAV p5 promoter. Plasmid pSA315 is analogous to pSA313 except that the CFTR cDNA is inserted in the reverse direction. Plasmid pSA306 is analogous to pSA315, except that it has a deletion of CFTR nucleotides 131 to 486. In both pSA315 and pSA306, the CFTR gene is expressed from the AAV ITR, as discussed below. The junction sequences between the CFTR insert and the AAV termini and SPA regions of pSA313, pSA315 and pSA306 were verified by DNA sequencing. pSA464 was derived from pSA306 by digestion with AfIII at one nucleotide of the CFTR sequence and filling in and blunt-end ligation with T4 DNA polymerase and T4 DNA ligase. This created a frameshift in the CFTR sequence. The presence of this mutation was verified by DNA sequencing.

Transfection: DNA transfection into IB3 was performed in 6- or 24-well dishes using lipofection. 30 µg of lipofection reagent (BRL, Gaithersburg, MD) was used for every 5 to 6 µg of transfected DNA. Lipofectin and DNA were mixed in 1.0 ml of serum-free LHC-8 medium and added to cells (5 x 105 to 5 x 106 in 35 mm wells) already covered with 0.5 ml of medium. Cells were exposed to DNA for 4 hours, rinsed with PBS, and then grown in 2 ml of fresh medium. DNA transfection into 293 cells was performed using the standard DNA calcium phosphate precipitation procedure.

Geneticin selection: IB3 cells used for stable neo expression were split 1:3 into 10 cm dishes 24 to 48 hours after transfection, and geneticin sulfate was added at 120 μl/ml 72 to 96 hours after transfection. The amount of geneticin used was based on a minimum lethal dose titration. Geneticin-resistant (genr) colonies were counted 14 to 16 days after the start of selection.

CFTR complementation: IB3 cells were plated at approximately 5 x 105 cells per 35 mm dish. Twenty-four hours after plating, cells were transfected by lipofection using 6 µg pAAVp5neo or 1 µg pAAVp5neo together with 5 µg pSA313, pSA315, pSA306 or pSA464 and geneticin selection was performed as described above. Genr colonies were isolated from each of the other two sets of plates 14 days after selection began. Each isolated colony was trypsinized using a cloning cylinder and expanded from 10 mm wells. After expansion of individual clones, cells were prepared for ³⁶Cl⁻ efflux assays and Western blot analysis.

Chloride efflux assays: Chloride efflux assays were performed as described (Trapnell et al., 1991, J. Biol. Chem. 266: 10319-10323) for individual clones at passage 4 to 8. Briefly, cells were grown in 35 mm dishes and loaded with 3 uCi 36Cl- in bicarbonate-free balanced Ringer's salt solution for 2 to 9 hours. Initial experiments involving repeated assays with the same clone of cells showed no significant differences in efflux after different loading times, and a loading time of 2 hours was then chosen for subsequent experiments. After loading, cells were washed 2 to 3 times with ice-cold 0.15 M NaCl, 5 mM Hepes, pH 7.4. 1 ml of Ringer's solution was added and immediately removed (time zero) and replaced with 1 ml of Ringer's solution. This process was repeated at various time points over a 15-minute period. The amount of radioactivity in each 1 ml sample of medium was determined by liquid scintillation counting. After the last sample was removed at time 15 minutes, the residual radioactivity remaining in the cells was determined by lysing the cells in 0.2 N NaOH and scintillation counting. The total radioactivity from all time points and the total cell lysate was then summed and the efflux was expressed as a percentage of the total radioactivity remaining in the cells at each time point. Efflux assays were then repeated for each clone examined using 10 µm forskolin dissolved in Ringer's efflux solution starting at time zero. The relative stimulation by forskolin was then calculated by calculating the rate (ka) of efflux in the presence of forskolin and expressed as a ratio relative to the rate of efflux in the absence of forskolin. For IB3 cells displaying the CF defect, this ratio is 1.0 or less. For cells complemented with CFTR vectors, this ratio is greater than 1.0.

Packaging of AAV2-CFTR vectors: Packaging of AAV2 vectors was accomplished by first infecting 293-31 cells (grown to semiconfluence in 100 mm dishes) with adenovirus type 5 (Ad5) (at a multiplicity of 5 to 10 infectious units/cell) and then cotransfecting the vector plasmid pSA306 or pSA464 (1 µg) and the packaging pAAV/Ad (5 µg) using the CaPO4 transfection procedure (Tratschin et al., 1984, Mol. Cell. Biol. 4: 2072-2081). Medium was replaced 2 hours before transfection and Ad5 was inoculated into the medium 1 hour before transfection. Medium was replaced 4 hours after transfection. Cells were cultured for 3 to 4 days and then harvested by gentle scraping into the medium. For direct analysis of the packet, lysates were frozen and thawed 3 times, debris was removed by low-speed centrifugation, and then warmed to 60ºC for 15 minutes to inactivate adenovirus. For use of vectors in transduction of IB3 cells, scraped cells were concentrated by low-speed centrifugation (4000 rpm) and resuspended in 10 mM Tris-HCl buffer, pH 8.0. Cells were lysed by freezing and thawing 3 times, and virus was concentrated and purified using CsCl density gradient ultracentrifugation (Carter et al., 1979, Virology 92: 449-462). Fractions used for transduction assays were then dialyzed against 1 x SSC three times for 1 hour at room temperature and heat treated at 60°C for 15 minutes to inactivate possible residual adenovirus. The titer of the vector preparation was determined by DNA slot blot hybridization (Samulski et al., 1989 J. Virol. 63: 3822-3828).

AAV2 particle-mediated transduction: Virus particle-mediated neo-transfection of IB3-1 CF bronchial epithelial cells was accomplished by infecting 103 to 4 x 104 cells in individual wells of a 24-well dish with a known number of AAV-CFTR vector particles per cell. Cells were grown for several weeks and assayed for complementation of the CF defect.

Expression of a gene from a promoter comprising only AAV ITR: In the course of constructing vectors designed to express CTFR from the AAV p5 promoter, we accidentally generated such a plasmid construct in which the gene was inserted in the reverse direction. This vector plasmid would not have been expected to function since it did not have a known promoter in the correct orientation. However, due to a fortunate coincidence in a laboratory experiment, we examined this plasmid construct and discovered that it fulfilled the function of expressing the gene. This prompted us to carefully examine the construct and we concluded that the ITR may act as a transcription promoter. As a result, we have carried out specific experiments, detailed in this application, which demonstrate that the ITR can act as a transcription promoter.

AAV vectors therefore only need to have the ITR sequences and a poly-A site to express a foreign gene. This is a new and novel finding and indeed contradicts the expectation based on previous teachings, where there was a general consensus that the AAV ITRs are not transcriptionally active (Walsh et al., 1992, PNAS (in press)). We show here that the AAV ITR is transcriptionally active in stable integration assays such that a functional CFTR cDNA is expressed.

Construction of AAV-CFTR vectors: Figure 1 shows the structure of several AAV-CFTR vectors designed to express CFTR either from the AAVp5 promoter, as in pSA313, or from the AAV ITR, as in pSA313 or pSA306. In pSA313, the CFTR cDNA (cross-hatched region and arrowhead) is inserted from 4500 nucleotides downstream of the AAV p5 promoter, i.e., AAV nucleotides 1 to 263, on the left. It contains the synthetic poly-A site. In pSA315, the CFTR cDNA was inserted in the opposite orientation so that it is downstream of the right AAV ITR sequence and the synthetic poly-A site. In this configuration, CFTR is expressed from the right ITR and the poly-A site can be read translationally throughout in the reverse direction, as noted above. In pSA306, the construct is exactly analogous to pSA313, except that 350 nucleotides of the amino-terminal region of CFTR cDNA (nucleotides 131 to 486) have been deleted. This results in expression from the right ITR of a fusion protein consisting of an N-terminally deleted CFTR protein with a fusion region at the N-terminus derived from reading the synthetic poly-A site throughout in the reverse direction, i.e., right to left in the orientation of Figure 1.

Plasmid pSA464 is a control derived from pSA306 by introducing a frameshift mutation at an AfIII site at nucleotide 993 so that no functional CFTR protein can be produced. This is indicated by a vertical outline bar.

Expression of CFTR and complementation of the CF defect in stable transfectants of CF airway cells: To examine the efficiency of the AAV-CFTR vectors for expression of the CFTR gene, the plasmids shown in Figure 1 were each transfected into IB3 cells together with pAAVp5neo using cationic liposomes (Lipofectin reagent, BRL, Gaithersburg, Md). Control cells were transfected with pAAVp5neo alone. Genr colonies were taken from the original plates and expanded into stable cultures and assayed for functional expression of the CFTR protein. All of these clones were stable with respect to neo expression during repeated passages over several months in culture.

Expression of CFTR can be demonstrated in functional assays in IB3 cells that display the cystic defect: a functional CFTR protein should restore Cl- conductance in these cells, which is regulated by cAMP and thus stimulated by forskolin (Drumm et al., 1990, Cell 62: 1227-1233; Hwang et al., 1989, Science 244: 1351-1353; Li et al., 1988, Nature (London) 331: 358-360; Li et al., 1989, Science 244: 1353-1356; Rich et al., 1990, Nature (London) 347: 358-363). Examples of Cl- efflux are shown in Figure 2 and a summary of the rate constants calculated from these data is shown in Figure 3. Both parental IB3 cells and the control N6 clone (transfected with pAAVp5neo alone) showed a relatively low rate of Cl- efflux that was unresponsive to forskolin (Figure 2). In contrast, a number of clones of the AAV-CFTR transfectants, as shown in Figure 2 for clones C35 and C38 (both derived from transfection with pSA306), showed significantly increased basal rates of efflux; more significantly, they showed the characteristic additional increase in efflux in response to forskolin. Efflux was measured in the absence ( ) or presence ( ) of 20 µM forskolin.

Figure 3 shows Cl- efflux assays in IB3-1 cells complemented with the CFTR gene by stable transfection of AAV-CFTR vectors. IB3 cells were transfected with pAAVp5neo and pSA313, pSA315, pSA306 or pSA464. Geneticin-resistant colonies were selected and analyzed for response to forskolin stimulation in a Cl- efflux assay. The ratio of the rate of efflux in the presence of forskolin to the rate in the absence of forskolin (ka (forskolin)/ka (Ringer's solution)) is plotted. For each vector, n denotes the number of individual clones that showed (hatched bars) or did not show (open bars) a forskolin response. For each group of clones, the mean ratio was calculated. For the parent IB3-1 cells or the cell clone transfected with pAAVp5neo alone, n denotes the number of measurements on the same clone.

Fig. 3 shows that 28% (4/14) of the pSA313 transfectants and 50% (6/12) of the transfectants with pSA315 or pSA306 were complemented for the defect. This indicates that all three vector constructs were functional. The increased number of functional clones with pSA313 or pSA306 may indicate that the ITR promoter in the vectors was more efficient than the p5 promoter in pSA313. None of the clones transfected with the control vector pSA464 were complemented. These results demonstrate two novel findings. First, the AAV ITR sequence also functions efficiently as a promoter upon stable integration into cells, as demonstrated by the function of both pSA313 and pSA306. Second, the truncated CFTR protein expressed from pSA306 is also functional in complementing the CFTR defect. In the pSA306 vector, the largest open reading frame expresses a fusion protein by reading most of the synthetic poly-A sequence through in reverse.

The observations with pSA306 are particularly important because it has been previously taught that the region of CFTR deleted in pSA306 is in fact essential for CFTR function when CFTR is expressed from various other vectors, such as vaccinia (Andersen et al., 1991, Science 251: 679-682). Furthermore, the overall size of the AAV-CFTR vector in pSA306 is equivalent to the size of wild-type AAV DNA, and thus this vector should be packageable into AAV particles for use as a transduction vector. We have investigated the packaging of the pSA306 vector into AAV particles. To study packaging of the AAV-CFTR vector pSA306 into AAV particles, adenovirus-infected 293 cells were transfected with the AAV-CFTR vector (pSA306) in the presence (+) or absence (-) of the AAV packaging plasmid (pAAV/Ad). Lysates of the cultures were prepared 72 hours after transfection and used to infect fresh cultures of adenovirus-infected 293 cells in the absence (- wild type) or presence (+ wild type) of added wild-type AAV particles (multiplicity of infection = 3). 40 hours post-infection, Hirt lysates of cells were prepared and viral DNA was electrophoresed in an agarose gel, transferred to nitrocellulose and hybridized with a CFTR-32P DNA probe specific for the SA306 vector (306) or with an AAV-32P DNA probe specific for wild-type AAV (AAV). Replication of the SA306 vector was only detected in lysates that had been packaged in the presence of pAAV/Ad and subsequently infected in the presence of added wild-type AAV particles, demonstrating that the AAV-CFTR vector could be packaged into transducing AAV particles.

To demonstrate the functionality of the transducing SA306-AAV-CFTR vector, IB3 cell cultures were transfected with vector preparations containing packaged SA306 or a SA464 control vector were infected at a multiplicity of approximately 300 to 400 vector particles per cell. Cultures were grown in culture for several weeks and examined for functional expression of CFTR. As shown in Figure 4, the culture infected with the SA306 vector (A0 cells) was functionally complemented for the CF defect as demonstrated by the response to forskolin. In contrast, the control culture infected with the SA464 control vector (2F2 cells) was not complemented as demonstrated by the lack of response to forskolin.

The results shown in Figures 2, 3 and 4 were confirmed by other functional assays including immunofluorescence detection of CFTR protein and electrophysiological assays using patch clamp techniques.

The results described above demonstrate complementation and stable correction of the CF defect in airway epithelial cells following cationic liposome-mediated transfection with AAV-CFTR vector or infection of the cells with AAV-CFTR vector transducing particles. These results demonstrate the utility of the AAV vectors and of the invention as implemented with AAV vectors using an ITR as a promoter and introducing a synthetic poly-A site with specific features.

Our studies with the AAV-CFTR vectors were performed as an initial step in evaluating the feasibility of using an AAV vector for gene therapy. In this regard, it is important that we have demonstrated stable complementation of the CF defect in cells derived from the bronchial epithelium, since this is the site of the main clinical manifestation of the disease and most likely the site to which gene therapy vectors will be directed. The complementation experiments with a retroviral vector (Drumm et al., 1990, Cell 62A: 1227-1233) reported were performed with CFPAC cells, which are pancreatic cells and not airway cells.

Expression of CFTR in vivo

AAV vectors, particularly those expressing a gene from the ITR, can be used to treat human patients in the following general manner. If the vector is to be delivered as transducing particles, it can first be packaged into AAV particles in the general manner described herein for the AAV-CFTR vector SA306 or using any other suitable packaging system. The transducing AAV vector can be purified to remove and/or inactivate any byproducts or toxic compounds by banding in CsCl or by any other suitable method. For AAV vectors expressing a functional CFTR gene or any other gene for treating a lung disease, the vector can be administered directly in vivo to the lungs either by intubation and bronchoscopy or by nebulizer or by nasal spray or by inhalation of a suitable formulation of nasal drops. For these or other diseases, the AAV vector particles can be administered in vivo by intravenous or enteric administration or perhaps subcutaneously.

The vector can also be used in ex vivo gene therapy procedures by removing cells from a patient, subsequently infecting them with the AAV vector particles, and reintroducing the cells into the patient after a period of maintenance and/or growth ex vivo.

The AAV vectors can also be administered in in vivo or ex vivo gene therapy procedures in various other formulations in which the vector plasmid is administered as free DNA either by direct injection or after incorporation into other delivery systems such as liposomes or systems designed to reach their target by receptor-mediated or other endocytic methods. The AAV vector can also be introduced into an adenovirus, retrovirus or other virus which can then be used as the delivery vehicle.

Other vectors using the promoter region sequences from ITR

An additional application of the present finding is the use of the ITR sequences responsible for promotion in other vectors. The IRT region of AAV does not have a normal TATA motif common to many eukaryotic promoters, and has not previously been recognized to act as a transcriptional promoter within the context of an AAV genome. It is likely that in the context of the AAV genome, this ITR does not act as a promoter, perhaps due to the actions of the other known AAV promoters downstream of it. However, not all eukaryotic transcriptional promoters require or possess the TATA motif. Having shown that the AAV ITR acts as a promoter, we examined the ITR sequence for elements likely to explain this function.

Examination of the ITR sequence reveals two motifs that are likely to be important in its function as a promoter. First, in the region between AAV nucleotides 125 and 145 (commonly known as the AAV-d sequence) there is the sequence 5'-AACTCCATCACT-3' [SEQ ID NO 1]. This sequence differs by only one base from similar sequences at the 5' start site of the promoters for the human terminal deoxynucleotidyl transferase gene and for the adenovirus major late gene promoter and agrees well with the consensus sequence for an element described as an Inr (initiator) element (S.T. Smale and D. Baltimore, 1989, Cell 57: 103-113; Smale et al., 1990, Proc. Natl. Acad. Sci. (USA) 87: 4509-4513). A second set of GC-rich elements is present in the ITR region between nucleotides 1 and 125, including the elements GGCCGCCCGGGC [SEQ ID NO 2] from nucleotide 41 to 50, AAAGCCCGGGCGTCGGGCGACC [SEQ ID NO 3] from nucleotide 51 to 73, GGTCGCCCGGCCTCA [SEQ ID NO 4] from nucleotide 76 to 90, and GAGCGGCGAGAG [SEQ ID NO 5] from nucleotide 101 to 112, which have strong homology to the set of consensus sites shown to be sites for the common transcription factor Sp1 (Pitluck and Ward, 1991, J. Virol. 65: 6661-6670). Finally, it is now known that an Inr sequence can function as a transcription promoter in the presence of sites for other factors, such as Sp1 (Smale and Baltimore, 1989; Smale et al., 1990).

It is likely that these or other regions of the ITR may be important in allowing it to function as a transcriptional promoter. It is now straightforward and obvious to others skilled in the field to perform standard mutagenesis techniques to alter the ITR sequence to precisely determine the control elements and modulate the transcriptional activity of the ITR up or down.

Sequence protocol

(1. General information:

(i) Applicant:

(A) Name: United States of America, represented by the Secretary, Department of Health

(B) Street: c/o National Institutes of Health

(C) City: Bethesda

(D) State: Maryland

(E) Country: USA

(F) Postal code: 20892

(ii) Title of the invention: Modified adeno-associated virus vector capable of expression from a new promoter

(iii) Number of sequences: 6

(iv) Machine-readable form:

(A) Type of media: diskette

(B) Computer: compatible with IBM PC

(C) Operating system: PC-DOS/MS-DOS

(D) Software: PatentIn Release #1.0, Version #1.30 (EPO)

(v) The present patent application:

Application number: EP 93916425.7

(2) Information on SEQ ID NO: 1:

(i) Sequence features:

(A) Length: 12 base pairs

(B) Type: Nucleic acid

(C) Strand texture: single

(D) Topology: linear

(ii) Molecular type: DNA (genomic)

(xi) Sequence description: SEQ ID NO: 1:

AACTCCATCA CT 12

(2) Information on SEQ ID NO: 2:

(i) Sequence features:

(A) Length: 12 base pairs

(B) Type: Nucleic acid

(C) Strand texture: single

(D) Topology: linear

(ii) Molecular type: DNA (genomic)

(xi) Sequence description: SEQ ID NO: 2:

GGCCGCCCGG GC 12

(2) Information on SEQ ID NO: 3:

(i) Sequence features:

(A) Length: 22 base pairs

(B) Type: Nucleic acid

(C) Strand texture: single

(D) Topology: linear

(ii) Molecular type: DNA (genomic)

(xi) Sequence description: SEQ ID NO: 3:

AAAGCCCGGG CGTCGGGCGA CC 22

(2) Information on SEQ ID NO: 4:

(i) Sequence features:

(A) Length: 15 base pairs

(B) Type: Nucleic acid

(C) Strand texture: single

(D) Topology: linear

(ii) Molecular type: DNA (genomic)

(xi) Sequence description: SEQ ID NO: 4:

GGTCGCCCGG CCTCA 15

(2) Information on SEQ ID NO: 5:

(i) Sequence features:

(A) Length: 12 base pairs

(B) Type: Nucleic acid

(C) Strand texture: single

(D) Topology: linear

(ii) Molecular type: DNA (genomic)

(xi) Sequence description: SEQ ID NO: 5:

GAGCGGCGAG AG 12

(2) Information on SEQ ID NO: 6:

(i) Sequence features:

(A) Length: 58 base pairs

(B) Type: Nucleic acid

(C) Strand texture: double

(D) Topology: linear

(ii) Molecular type: DNA (genomic)

(xi) Sequence description: SEQ ID NO: 6:

Claims (10)

1. Functional cystic fibrosis transmembrane trafficking regulator protein with a deletion of any or all of the amino-terminal 118 amino acids. 2. Protein according to claim 1, wherein the deletion affects amino acids 1 to 118. 3. A polynucleotide comprising the inverted terminal repeat sequences (ITR sequences) of adeno-associated virus and a heterologous nucleic acid encoding the protein of claim 1 or claim 2, wherein the ITR sequences promote transcription of the nucleic acid. 4. Polynucleotide according to claim 3, also comprising a polyA site having the nucleotide sequence: 5. A vector comprising the polynucleotide of claim 3 or claim 4. 6. Vector according to claim 5, which is an adeno-associated virus vector. 7. A host cell containing the vector according to claim 5 or claim 6. 8. The host cell of claim 7, which is an epithelial cell. 9. A composition comprising the vector according to claim 5 or claim 6 and a pharmaceutically acceptable carrier. 10. Use of the vector of claim 5 or claim 6 for the manufacture of a medicament for use in the treatment of cystic fibrosis in an individual.